Systemic factors

Macular edema manifests clinically as retinal thickening that results from chronic breakdown of the blood–retinal barrier. Extracellular fluid accumulates in the outer plexiform layer and extends into the inner retina with increasing severity. The pathophysiologic factors that govern the formation of edema in the macula are similar to those that contribute to edema elsewhere, such as in the brain. That is, tissue edema occurs when the ability of the blood vessels and the surrounding tissue to regulate leakage are overcome either by increased intralumenal hydrostatic pressure or increased extravascular (tissue) concentration of albumin. This relationship is classically described in Starling’s Law of the Capillary, which states that fluid filtration is related to the forces that drive fluid across the vessel wall (blood pressure, intravascular fluid volume, and interstitial fluid osmotic pressure) minus the forces that promote fluid reabsorption (plasma oncotic pressure). The importance of this concept has been related to macular edema in diabetes by Kristinsson and co-workers¹. This physiologic principle is useful to guide investigations into macular edema and the evaluation and treatment of patients.

The primary systemic factor that contributes to disruption of the normal autoregulatory process of the retinal vasculature is systemic arterial hypertension. Epidemiologic studie have confirmed that hypertension aggravates existing retinopathy and increases the risk of developing retinopathy ². This point is illustrated in Figure 62.1. The patient whose eye is illustrated is a 65-year-old woman with 10 years of type 2 diabetes, and her blood pressure was 150/105 mmHg. The eye has diffusely narrowed retinal arterioles, cotton wool spots, and hemorrhages. The macula is thickened, and lipid exudates threaten the fovea, and visual acuity was 20/40. The other eye had a similar appearance.

Fig. 62.1 Effect of uncontrolled hypertension on diabetic macular edema. The right eye of a 65-year-old female with 10 years of type II diabetes whose blood pressure was 150/105 mmHg. Note the diffusely narrowed retinal arterioles, the nerve fiber layer infarcts and hemorrhages. Her right macula is thickened with lipid exudates threatening the fovea, and her visual acuity was 20/40. The macular edema did not respond well to focal laser because her blood pressure remained elevated.

In addition to hypertension, intravascular fluid overload from congestive heart failure or renal failure increases the hydrostatic pressure across the vascular wall and the propensity to edema formation. The ophthalmologist should ask patients with diabetic macular edema about symptoms of fluid overload, such as orthopnea and dyspnea, and should observe whether those patients have pitting edema of their ankles.

Patients with hyperlipidemia may also have an increased tendency for macular edema formation when serum lipoproteins accumulate in the retina and exert the osmotic influences that draw water out of retinal vessels and cause retinal thickening. Data from the Early Treatment Diabetic Retinopathy Study (ETDRS) showed a twofold increase in risk of macular edema in visual loss in patients with total serum cholesterol greater than 240 mg/dl³.

The issue of glycemic control as an isolated factor in the development of macular edema is difficult to determine because insulin affects numerous aspects of cellular metabolism. However, in the Diabetes Control and Complications Trial (DCCT), the risk of macular edema was approximately 50% greater in the group that received conventional treatment than in the group under intensive treatment. The overall risk of developing macular edema in the intensively treated patients in the secondary intervention group (those patients who had mild to moderate retinopathy at the beginning of the study) was 27%⁴.

Diabetic nephropathy is also associated with increased incidence and progression of retinopathy in general and macular edema in particular. This correlation of diabetic nephropathy and retinopathy is known as the diabetic ‘renal–retinal syndrome’. The risk of macular edema is greatest in patients with overt proteinuria⁵, and systolic hypertension and higher glycosylated hemoglobin are also important factors in the Wisconsin Epidemiologic Study of Diabetes⁶. Nephropathy may aggravate macular edema via hypertension, fluid overload, hypoproteinemia, hyperlipidemia, or increased difficulty with metabolic control. Diabetic macular edema is a risk factor for increased mortality from heart disease⁷.

Several reports have suggested that thiazolidinedione (glitazone) drugs that are used widely as effective agents for type 2 diabetes may increase the risk of diabetic macular edema^8,9. Therefore, patients should be questioned about the use of these medications if they have a recent onset of DME associated with systemic edema but without cardiac or renal failure.

Taken together, poor metabolic (glycemic) control, increased blood pressure and serum lipids, nephropathy, and some medications contribute to the development of retinopathy and macular edema.

Ocular risk factors

In addition to systemic influences on the retinal circulation, several ocular conditions may also increase the risk of macular edema.

The combination of cataract and diabetic retinopathy is a common and clinically challenging problem. Cataract extraction, even with an intact posterior capsule and a posterior chamber intraocular lens, frequently increases the severity of retinopathy and macular edema⁷. The reasons for this are unclear, but may relate to release of inflammatory mediators in the perioperative period. The risk of macular edema after cataract extraction increases with increasing severity of overall retinopathy, and with increasing age of the patients. In a series of 109 patients, Benson and co-workers¹⁰ found that only 48% of patients undergoing extracapsular cataract extraction achieved 20/40 or better final acuity, and 28% tested at 20/200 or less. Only 65% improved two or more lines of acuity. A more recent study¹¹ showed that cataract surgery accelerates the overall progression of retinopathy after 1 year.

An additional risk factor for the development of macular edema is intensive panretinal photocoagulation¹². Macular edema is particularly likely to be increased after panretinal photocoagulation if the treatment is applied in a single session or close to the macula. Avoidance of the macular region during laser surgery may reduce this risk, and this adverse event appears to be much less common now than when reported initially.

Contraction of the posterior hyaloid face of the vitreous can exert traction on the macula, and this traction may contribute to the formation of macular edema, even in eyes that have received focal laser treatment¹³.

Systemic treatment

The primary factor to evaluate is metabolic control. As mentioned above, the risk of macular edema was significantly lower in the intensively treated group than in the conventional treatment group in the DCCT. However, no study has directly addressed the effects of improving metabolic control in patients with existing macular edema. In general, glycemic levels should be as low as can be achieved and tolerated by the patient without undue risk of hypoglycemia. A hemoglobin A₁C value less than or equal to 6% is a desirable goal, even if difficult for many patients to attain. In patients who have poor metabolic control, it is generally advisable to improve the metabolic control gradually over 6–12 months, because rapid tightening of control can cause transient worsening of retinopathy, usually made manifest by increasing ‘cotton wool’ spots or intraretinal microvascular abnormalities; rarely, neovascularization develops. In the DCCT transient worsening occurred in 22% of the intensively treated group and 13% of the conventionally treated group¹⁴.

Blood pressure in persons with diabetes should be 130/80 mmHg or less¹⁵. Blood pressure control significantly lowers the risk of progression of incipient diabetic nephropathy to overt nephropathy, and is reasonable to apply in patients with retinopathy as well. The United Kingdom Prospective Diabetes Study (UKPDS) showed that blood pressure control reduced the risk of retinopathy progression by one-third¹⁶. Hypertension is also associated with increased risk of retinopathy, even in children and young adults with type 1 diabetes¹⁷, although there is currently no evidence that proves blood pressure reduction in patients with type 1 diabetes reduces the risk of retinopathy. However, treatment with the renin angiotensin system inhibitors losartan and enalapril reduces the risk of retinopathy (but not nephropathy) independently of blood pressure¹⁸.

Serum lipids should also be evaluated and treated in accordance with guidelines for the reduction of stroke and myocardial infarction¹⁹. LDL cholesterol should be less than 70–100 mg/dl and non-HDL <100 mg/dl. The availability of improved medications for the control of hyperglycemia, hyperlipidemia, and hypertension over the last several years makes this possible. Identification and treatment of systemic risk factors can have substantial benefits in macular edema, even that which is clinically significant, as demonstrated in Figure 62.2. By identifying these systemic risk factors, the ophthalmologist has the opportunity to improve the visual prognosis, and the prognosis for life as well. To have this impact, the ophthalmologist must view diabetic macular edema as a systemic as well as ocular disease. This approach can facilitate collaboration with primary care physicians, diabetologists, and other specialists involved in the care of persons with diabetes.

Fig. 62.2 Effect of improved metabolic control on diffuse macular edema. (A) The right eye of a 47-year-old woman with poorly controlled type II diabetes (hemoglobin A₁C 12.5%), hyperlipidemia (total cholesterol 250 mg/dl), and triglycerides (542 mg/dl). Visual acuity was 20/60. Note the diffuse macular lipid exudates and thickening. (B) The same eye 9 months later when the patient’s cholesterol level was 200 mg/dl, her triglycerides were 240 mg/dl, and her hemoglobin A₁C was 10.2%. The macular lipid exudates and thickening have almost completely regressed without photocoagulation and visual acuity is 20/30.

Ocular treatment

The ETDRS¹² established the role of macular photocoagulation for patients with ‘clinically significant macular edema’. Clinically significant macular edema is defined as ‘retinal thickening at or within 500 μm of the center of the macula, hard exudate at or within 500 μm of the center of the macula associated with retinal thickening, or retinal thickening one disc area or larger within one disc diameter of the center of the macula’. Focal laser surgery reduces the risk of moderate visual loss (doubling of the visual angle) by approximately 50%. The benefit of this treatment was independent of the initial visual acuity, from 20/20 to 20/200. Treatment involves direct ablation of microaneurysms with or without a ‘grid’ pattern to areas of retinal thickening. Focal treatment is more effective at reducing macular thickening than is grid treatment and is the preferred approach²⁰. Foveal ischemia may coexist with macular edema and contribute to visual loss and, because it is associated with a worse visual prognosis, a preoperative fluorescein angiogram helps in the assessment of the foveal circulation. Patients with mild irregularity or enlargement of the foveal avascular zone may benefit from macular photocoagulation, but those with highly irregular avascular zones or greater than one disc diameter of foveal ischemia often will have less benefit, and their vision may be more likely to deteriorate after photocoagulation.

For focal treatment of diabetic macular edema, the spot size ranges from 50 to 100 μm in diameter, and the burns should cause light blanching of the microaneurysms or underlying pigment epithelium²¹. The power setting will necessarily be higher in patients with hazy media or light fundus pigmentation. The required power may also vary with the degree of retinal thickening. For example, a burn duration of 0.1 second usually allows an adequate burn and is short enough to prevent spread of the lesion if the patient moves the eye, but if the retina is thickening the laser energy is scattered, and higher power or longer duration exposures may be required. The wavelength of the laser light has not been shown to result in clinically important differences in results, despite theoretical considerations of absorption of blue-green light by macular xanthophyll pigment. The ophthalmologist should try to treat all of the detectable sites of leakage but avoid involvement of the foveal avascular zone. Two to three months may be required before it is possible to determine the response of an eye to macular photocoagulation, and if clinically significant thickening remains after 3–6 months, then additional treatment is probably indicated. A recent clinical trial comparing focal laser with intravitreal triamcinolone showed that the effects of focal laser on visual acuity and macular thickness continued to accrue over 2 years²².

In eyes that have coexisting proliferative diabetic retinopathy or severe non-proliferative retinopathy, panretinal photocoagulation may be applied concurrently. If the risk of immediate severe visual loss from vitreous hemorrhage appears low, then it is probably reasonable to apply macular photocoagulation first. Eyes that have vitreous hemorrhage or active neovascularization, however, may be treated with concurrent panretinal and macular photocoagulation. Panretinal photocoagulation improves retinal vascular autoregulation²³ and macular edema in selected cases, particularly those with florid neovascularization^24,25. Panretinal photocoagulation applied initially in the midperiphery, and then to the more posterior retina if needed, is associated with a lower risk of macular edema than panretinal photocoagulation applied to the entire retina simultaneously²⁶.

This discussion points out that diabetic macular edema is often multifactorial in origin, so effective treatment should include consideration of both systemic and ocular factors.

Safe and effective pharmacotherapy for DME has been a goal for many years. The approaches have included drugs with pleiotropic actions on inflammation and vascular integrity such as corticosteroids, and targeted agents directed at single molecules. Steroids induce endothelial cell tight junction protein expression^27,28 and reduce intravitreal inflammatory molecules²⁹. Numerous case series suggested that intravitreal triamcinolone could improve diabetic macular edema, and one controlled clinical trial³⁰ supported this conclusion. However, in the patients studied in the Diabetic Retinopathy Clinical Research Network trial, triamcinolone was inferior to focal photocoagulation²². Moreover, repeated intravitreal corticosteroid injections carry a very high risk of cataract (54%) and glaucoma (44%)^30,31. Intravitreal dexamethasone implants have benefits after 6 months of treatment³² and longer-term results are pending.

Vascular endothelial growth factor (VEGF) is a potent endothelial cell permeabilizing agent that disrupts endothelial tight junctions³³. Recent studies have demonstrated a moderate improvement of visual acuity in eyes treated with ranibizumab or bevacizumab^34,35. As of this date (August, 2010) no benefit on DME has been shown to result from inhibiting inflammatory cytokines such as tumor necrosis factor.

In summary, the pathogenesis and treatment of DME involve both ocular and systemic factors, all of which should be evaluated for optimal outcomes for our patients. Focal laser treatment remains the standard of care for DME, although intravitreal VEGF antagonists and/or steroids may add marginal benefits. Clearly, much more work remains to be done to reduce the risk of this vision-robbing thief.